Supported Interactive Electrocatalysts for Hydrogen

Journal of New Materials for Electrochemical Systems 9, 83-97 (2006) © J. New Mat. Electrochem. Systems

Supported Interactive Electrocatalysts for Hydrogen and Oxygen Electrode Reactions† Nedeljko V. Krstajic1, Ljiljana M. Vracar1, Stelios G. Neophytides2, Jelena M. Jaksic1,2, Kuniaki Murase3, Reidar Tunold4 and *Milan M. Jaksic1,2 1

Faculty of Technology and Metallurgy, University of Belgrade, Belgrade, Serbia & Montenegro 2 Institute of Chemical Engineering and High Temperature Chemical Processes, FORTH, and Department of Chemistry, University of Patras, 26500 Patras, Greece 3

Department of Materials Science and Engineering, Kyoto University, Kyoto, Japan

4

Department of Materials Technology, Norwegian University of Science and Technology, Trondheim, Norway

Received: September 19, 2005, Accepted: March 10, 2006 Abstract: Magneli phases have been introduced as an unique interactive support for electrocatalysts in hydrogen (HELR) and oxygen (OELR) electrode reactions both in water electrolysis and Low Temperature PEM Fuel Cells (LT PEM FC). The Strong Metal-Support Interction (SMSI) implies (a) the hypo-hyper-d-d-interbonding effect and (b) the interactive primary oxide (M-OH) spillover from hypo-d-oxide support. The stronger the bonding, the less strong the intermedaite adsorptive strength in the rate determining step (RDS), and consequently the faster arises the facilitated catalytic electrode reaction, while the primary oxide directly interferes and reacts either individually and directly to finsih oxygen reduction, or with such interactive species (CO) to contribute to CO tolerance. In such a context, the monoatimic network dispersion of Pt upon Magneli phases contributed to produce an advanced interactive supported electrocatalyst for cathodic oxygen reduction (ORR). The conditions to provide Au to act as the reversible hydrogen electrode have been proved either by its potentiodynamic surface reconstruction in a heavy water solution, or by the nanostructured SMSI Au on anatase titania with characteristic strained dorbitals in such an hypo-hyper-d-interactive bonding (Au/TiO2). In the same interactive boning sense, the thermodynamic basis and conditions to obtain the Brewer type symmetric Laves hypo-hyper-d-intermetallic phases (TiPt3, ZrPt3, HfPt3) by thermal reduction of their stoichiometrically mixed oxides, hydroxides or other easy decomposing salts in hydrogen furnace, have also been given and illustrated. Keywords: Magneli phases, SMSI (Strong Metal-Support Interaction), nanostructured monoatomic network dispersion, synergistic metal/support interactive electrocatalytic effect, the strained d-orbitals catalytic effect, hypo-hyper-d-inerelectronic bonding, Laves hypo-hyper-d-interatomic phases, reversible H2/Au/TiO2 electrode.

the s,p-electrons contribute a constant term to both of them all along the Periodic Table [6-9]. In such a context, the individual polatization or electrocatalytic properties for the cathodic hydrogen evolution (HER) feature periodic volcano dependence along thee transition series with maximum in the activity at the highest d-orbital exposures (Ni, Pd, Pt) or at d8-electronic configuration as an optimum [10,11]. Since the Brewer intermetallic bonding theory [8-9] established an interdependence between the average electronic configuration

1.0. Introduction The d-band is both the boning and chemisorptive orbital with linear interdependence between surface free- and cohesive- energy (in so called «one by six relation») [1-5], while †

Paper presented at 3rd Goerischer Memorial on Electrocatalysis, Berlin, Germany, July 6 — 8, 2005. *To whom correspondence should be addressed: Email: [email protected] , Mailing address: ICEHT/FORTH, P. O. Box 1414, Stadiou Road, Platani, GR-265 00 Patras, Greece

83

84

Nedeljko V. Krstajic et al. / J. New Mat. Electrochem. Systems

Figure. 1. Electrocatalytic activities of various intermetallic phases along Zr — Ni phase diagram (curve 1) for the HER, taken as the exchanged current (jo, close circles, curve 3) and relative current density changes (j, closed circuits, curve 2) at constant overvoltage ( - 0.2 V).

and crystal structure, the same periodic relation in electrocatalytic features has been found for all hypo-hyper-dinterelectronic combinations of transition metals with an optimum for the face centered cubic system (fcc) [3,5,12]. In other words, every hypo-d-electronic metal in interelectronic combination with hyper-d-elements imposes the interrelating volcano plots both in the bonding effectiveness and electrocatalytic activity for the HER along such an antibonding semi-series, and vice versa. In such a respect, there has been already shown that every hypo-hyper-d-phase diagram behaves as the part of the Periodic Table between two periods of initial free constituents [3,12]. Such state of experimental evidence that thermal stability peaks, as a measure of cohesive bonding effectiveness, follow the same type of curve for the electrocatalytic activity in the HER along all hypo-hyper-d-interelectronic phases diagrams (Fig. 1), means the established linck between the Brewer intermetallic bonding model and the theory of electrocatalysis for hydrogen electrode reactions (HELR), both cathodic HER and anodic H-adatoms oxidation (HOR) [4,5,12]. The same interionic bonding effectiveness, in the sense of extended Brewer theory [3,5], characterises the so called Strong Metal-Supports Interaction (SMSI) [13-18], whenever some pure or prevailing hyper-d-electronic transition metal catalyst arises supported by a hypo-d-oxide. The stronger the bonding, the less strong arises the adsorptive bond of Hadatom intermediates, and consequently (the Sabatier principle in heterogeneous catalysis [19]), the higher appears to be catalytic and electrocatalytic activity for the HER. For

larger nanostructured metallic catalysts such an advanced catalytic effect arises to be effective only around the circumference at the interphase M/TiO2. However, when the metallic particle size tends to monoatomic dispersion, the SMSI approaches its maximum in its interactive catalytic activity contribution. Thus, the main aim of the present paper is to introduce specific novel type of supported individual or composite nanostructured hypo-hyper-d-intermetallic phases of transition elements and interionic electrocatalysts deposited upon Magneli phases or the modified surface structure of the latter, as the interactive and highly conductive oxide (TinO(2n-1) in general, or Ti4O7 in average) carriers with corresponding improvements in their synergistic catalytic activity, CO tolerance, advanced electronic conductivity and current collection. In fact, the overall target would be to combine and advance the synergistic electrocatalytic and oxophilic primary oxide (M-OH) spillover contribution to CO tolerance in the anodic hydrogen (H-adatom) oxidation, cathodic oxygen reduction (ORR) and similar electrode processes, both by the intermetallic and SMSI on the basis of hypo-hyper-d-interelectronic, basically interionic, interactive bonding effects.

1.1. Interactive Spillover Effect of Catalyst Supports There is a strong first principle thermodynamic confirmation (Density Functional Calculations, DFC) [20-22], that water molecules undergo spontaneous dissociative adsorption on anatase and even rutile titania surfaces. More specifi-

Supported Interactive Electrocatalysts for Hydrogen and Oxygen Electrode Reactions / J. New Mat. Electrochem. Systems

85

Scheme I. Model presentation for the interactive SMSI effect resulting in the spillover transfer of the primary oxide (MOH) as a dipole from hypo-d-oxide upon metallic part (fcc Au) of such a composite electrocatalyst. Figure. 2. The perspective views of DFT-optimised atomic structures for: (a) the clean anatase (ADM) ad-molecule model of unreconstructed (001) surface; (b) the dissociated state of water (0.5 monolayer) on (001); (c) the relaxed geometries of molecular state of adsorbed water (1.0 monolayer of hydroxylated anatase) on (001); and (d) the mixed state of water on (001) with a half-dissociated coverage of adsorbed monolayer water molecules (courtesy by A. Vittadini [20,21]).

cally, the (101) surface of anatase characterizes molecular adsorption with partial dissociation (less than 50%), while at (001) surface proceeds spontaneous and prevailing dissociative chemisorption of water molecules (Fig. 2) [20,21]. In addition, there has recently been shown by the firstprinciples molecular-dynamic simulations the existence of a mechanism for thermodynamically favored spontaneous dissociation of water at low coverage at oxygen vacancies of the anatase (101) surface, and consequently to the Magneli phases as substantionally suboxide structure, with respect to molecular adsorption [22]. This is the status of reversible open circuit dissociative adsorption of water molecules at the equilibrium state, while in the presence of metallic SMSI part of catalyst, directional electric field (or, electrode polarization) further disturbs such an established equilibria and dynamically imposes further continuous dissociation of water molecules. More specifically, hypo-d-electronic transition metal ions feature several altervalent states giving rise even to mixed valence compounds (TiO2/WO3), so that such hydrous oxide network, in particular of polyvalent (high altervalent capacity) hypo-d-electronic elements, substantially behaves as ion exchange membrane [4,5,23]. In fact, gels (aero and xerogels) are biphasic systems in which solvent (water) molecules are trapped inside an oxide network, and such material can be considered as a water-oxide membrane composite [5,23]. Water molecules are adsorbed at the surface of the oxide particles within such a supporting oxide

Pt3Ti

PtO

Pt(CxHyCl) 0.7 eV

75.7 eV

XPS Intensity / a.u.

73.6 eV 71.6 eV Pt foil

Pt3Ti

78 76 74 72 70 68 66

(1% Pt/TiO2)

84

82

80

78

76

74

72

70

68

66

64

Binding Energy / eV Figure. 3. The XPS Pt 4f spectra scanned at the interphase between Pt catalyst (1.0 wt.%) and anatase TiO2 support in their SMSI. Deconvolution exactly reveals the existence of the TiPt3 intermetallic phase with all identified spectral properties.

network and consequently, such composite material exhibits specific properties arising from the intimate mixing and interacting (by adsorptive dissociation of water moleculs, Fig.2 [20-22]) of both phases. The main consequences of substantial significance are (a) continuous undisturbed reversible membrane mechanism of the altervalent changes (Ti3+ ⇔ Ti4+) [4,5]:

Σ

Ti(OH)4 + M → Ti(OH)3+ + M-OH + e-

(1a)

Ti(OH)3+ + 2H2O → Ti(OH)4 + H3O+

(2b)

M + 2H2O ↔ M-OH + H3O + e +

-

(1)

resulting in OH- - transfer within such an ion exchange membrane, with consequent spillover of primary oxide (M-


86

Cyclic Voltam m etry Data for a Range of Materials 0.2

0.15

0.1 0.05 Ebonex

0 -0.5

0

0.5

1

1.5

2

2.5

3

Resinf illed carbon Por ous carbon

-0.05

Elect rogr aphit e

-0.1 - 0.15

-0.2

-0.25 Vol t a ge v s. M e r c ur y - M e r c ur y S ul f a t e

Figure. 4. Potentiodynamic presentation of the polarization properties of Magneli phases (Ebonex), and therefrom their chemical stability, as compared with (a) resin filled carbon, (b) porous carbon and (c) electrographite.

Scheme II. The model of interactive grafting of Pt-acac (acethylacetonate, Pt-2,4-pentaneditionate) upon titania support.

OH), over metallic catalyst particles (Scheme I), and (b) the spontaneous reduction of TiO2 by chemisorbed H-adatoms (M-H) to its suboxide state (TiOx, 2 > x > 1) [16-18], the latter leading at higher temperature either to interphase formation of stable intermetallic phases (TiPt3) (Fig. 3), or to the highly stable non-stoichiometric Magneli phases themselves. In such a respect, XPS analysis of sol-gel formed submonolayered anatase titania upon Vulcan XC-72, after thermal decomposition of Pt-acac (acetylacetonate or 2,4pentandionate) to get nanostructured Pt/TiO2, shows no charge separation and consequently, electron conductivity of such a type electrocatalysts . In addition, such a thermodynamically predestined structure of anatase titania for spontaneously dissociative adsorption of water molecules, more than Magneli phases, affords the proper interactive substrate for the grafting adsorption of individual and prevailing hyper-d-electronic nanostructured precursors (Scheme II). The adsorbing substrate features positively charged properties vs. p.z.c., while the grafting

species has a negative overall colloidal charge. In conjunction with such property, it would be worthwhile mentioning that properly oxidized carbon supports, upon which surface appear orderly and homogeneously distributed interactive C-OH groups [24], behave as predestined centers for grafting of colloidal catalytic precursors, on the same way as above discussed and displayed the spontaneously dissociatively adsorbed OH-groups on anatase or Magneli phases themselves. In such a context, the present paper aim has been to provide (a) highly electron-conductive, (b) strong hypo-hyper-dinterelectron bonding and consequently, less H-adatom adsorptive strengthened intermediates, (c) advanced spillover interactive hypo-d-oxide supports for primary oxides (M-OH) transference and (d) a priori ready for an ordered monoatomic network dispersion in individual or prevailing hyper-delectronic catalyst grafting. As an extremely orderly distributed layerlike structure, in particular for suboxide centers network distribution, Magneli phases of rather high chemical stability and almost metallic type conductivity, appear as such an unique and highly advanced catalytic interactive support. 2.0. Experimental Magneli phases of non-stoichiometric titania or titanium suboxide (general composition TinO2n-1, 10 ≥ n ≥ 4, in average Ti4O7, Ebonex, Atraveda Ltd., Mansfield, U.K.), with advanced conductivity (300 - 1,000 S cm-1), particle size bellow 10 μm, or ground in planetary mills for one order of smaller size, have been the main catalytic support for the very first time introduced in electrocatalysis, either plain or with submonolayer of anatase titania (TiO2) and/or in admixture of peroxopolytungstic acid (PTA) with the latter. Magneli phases are the most advantageous both current collecting and electron conducting support (SMSI), but of much smaller


available surface area (0.2 — 1.2 m2 g-1) relative to anatase (∼ 250 m2 g-1), and as it has been shown in the present study, the hypo-d-suboxide with unique ordered interactive catalyst grafting and spillover properties in the primary oxide transfer. Magneli phase stability illustrates its cyclic voltammogram (Fig. 4). Majority of experiments have been carried out in sol-gel synthesis [4,5]. First of all, anatase titania as the main composite hypo-d-oxide catalytic support has been produced by hydrolysis of Ti-isopropoxide in anhydrous ethanol and spread as a submonolayer upon Vulcan XC-72 carbon or Magneli phases themselves, and after supercritical drying with CO2, calcinated in air at 300 — 400 oC to develop available surface area between 180 and 250 m2 g-1. The sol-gel process to obtain 10 nm-sized peroxopolytungstic acid (PTA) sol was also adapted [4,5]. Such an ethanol solution of homogeneous colloidal particle size of PTA was then used in admixture with Ti-isopropoxide to get a mutual polymeric network of about 3 to 7 mol % PTA, calculated on the basis of WO3, that has been the main submonolayer composite hypo-d-electronic oxide support of Magneli phases for the SMSI and this way increased surface area for grafting catalysts. On such a manner, the altervalent capacity of titania (z = 4), has been increased to higher value (z = 6), for much advanced primary oxide (M-OH) spillover. In other words, Magneli phases have been the main interacting and grafting individual support of highly advanced conductivity, while its submonolayer titania, or the latter in combination with nanolayered PTA, were further advanced catalytic SMSI carriers of this way increased surface area for higher subnanostructured network dispersion of metallic electrocatalysts. The initial Magneli phase (Ebonex) powder was ground below 100 mesh by using a Fritsch Planetary Mill (Pulverisisette 5), with ball milling (6 mm diam) performed with maximum acceleration of balls of 700 m2 s-1 in an argon atmosphere. Nitrogen adsorption and desorption isotherm, that presents the amount of N2 adsorbed as a function of relative pressure at —196 °C, before addition of nanosized Pt particles, was done and specific surface area of catalyst support was calculated by BET equation. SBET was determined as 1.6 m2 g-1 and could be classified as very low. Ebonex support was characterized by recording their powder X-ray diffraction (XRD) patterns on a Siemens D500 Xray diffractometer using Cu Kα radiation with a Ni filter. The 2θ angular regions between 20 and 850 were explored at scan rate of 0.020/s with the angular resolution of 0.020 for all XRD tests. The chemical composition of Ebonex was determined by energy dispersive X-ray spectroscopy (EDX), too. Pt catalyst for the ORR was prepared by a conventional impregnation method using 2-propanol solution of Pt(NH3)2(NO2)2 (Merck) and Ebonex powder as supporting material. The preparation process can be described as follows: Appopriate amount of Pt(NH3)2(NO2)2 solution (10 mg/ml Pt) was slowly added to 90 mg of Ebonex . This was ultrasonically blended for 1 h and after thorough mixing, the

87

precursor suspension was allowed to dry at 80 0C for 12 h. The precursor powder was then set in a tube furnace and reduced under flowing H2 gas at 300 0C for 2 hours and cooled at room temperature under flowing inert (Ar) gas. The Pt loading of the all catalysts was always 10 wt.%. Five milligram of Pt/(Magneli phases) catalyst was ultrasonically suspended in 1.0 ml of water-methanol (v/v = 1/1) and 50 μl of Nafion solution (5 wt.% Aldrich solution) to prepare catalyst ink. Then, 10.0 μl of ink was transferred with an injector to clean gold disk electrode (4 mm diameter, with area of 0.125 cm2). After the water-methanol volatilization, the electrode was heated at 80 0C for 10 min. The Pt loading of the all catalysts was always 5.4 μg. Transmission electron microscopy (TEM) measurements were performed using the FEI (Fillips electronic instruments) — CM200 super-twin and CM300 ultra-twin microscopes operating at 200 and 300 kV and equipped with the Gatan 1kx1k and 2kx2k CCD cameras, respectively. Specimens were prepared for transmission electron microscopy by making suspensions of the catalyst powder in ethanol, using an ultrasonic bath. These suspensions were dropped onto clean holey carbon grids, which were then dried in air. A reversible hydrogen electrode (RHE) in the same solution as that of the working electrode was used as the reference electrode. A large-area platinim sheet of 5 cm2 geometric area was used as the counter electrode. The electrochemical measurements were performed in 0.5 M HClO4 solution ( Spectrograde, Merck), prepared in deionized water (“Millipore”- 18 MΩ), at the temperature of 20.0 °C. The cyclic voltammetry experiments were performed in the potential range between hydrogen and oxygen evolution with a various scan rates at a rotating speed of 2500 rpm, at Au, Au/Ebonex and Au/Ebonex/Pt electrodes. Some specific potentiodynamic scans were undertaken in alkaline (0.1 M NaOD) heavy water solution with spectroscopically pure stationary Au electrode. For Au/TiO2 interaction, rough, high surface area nanocrystalline anatase titania thin films were prepared on microscopy glass slides applying the doctor-blade procedure [25,26]. Gold films were deposited by electron beam evaporation of an Au wire of ultrahigh purity under vacuum higher than 6 x 10-4 Pa, onto stationary, titania coated microscopical slides, placed in parallel to the emitting surface. Quartz crystal monitor was used for controlling the deposition rate of 0.01 μg cm-2s-1. To elucidate the gold valence state and Au/TiO2 binding energy, X-ray photoelectron spectroscopy (XPS) experiments were carried out in an ultrahigh vacuum chamber equiped with a hemispherical electron analyzer (SPECS LH-10) and twin-anode X-ray source, using unmonochromatized Mg-Kα at 1253.6 eV and an analyzer pass energy of 97 eV. Other details of experimental procedures, in particular as concerns apparatus and methods of measurements, were displayed in our preceding papers [4,5,27]. 3.0. Results and Discussion 3.1. Magneli Phase Interactive Supported Pt as

88


Figure. 5. TEM images of Pt nanoparticles monoatomically dispersed upon Magneli phases (Ebonex, Atraverda) for low and hgh magnification.

Figure. 6. Tafel plots for the cathodic ORR scanned on monoatomic dispersed Pt (5.4 µm cm-2) upon Magneli phases (open circuits) and nanostructured Pt (10 wt.% Pt on Vulcan XC-72, triangles) in 0.5 M HClO4 solution (backwards direction scan).

Advanced Electrocatalyst for Cathodic Oxygen Reduction (ORR) The most outstanding electrocatalytic effect, substantially advanced by the interactive Magneli phases support has recently been substantiated by the extremely highly dispersed Pt sub-nanostructured monoatomic particle network. Such Pt deposit of rather unusual, as high as 36 m2g-1 Pt upon a rather small (1.6 m2g-1) available Magneli phase surface, was obtained from 2-propanol solution of

Pt(NH3)2(NO2)2 by the interactive grafting deposition followed by reduction in hydrogen stream at 300 oC [28]. High resolution TEM analysis (Fig. 5) shows particle distribution between 0.1 and 1.0 nm, with prevailing monoatomic dispersion, though in average only every 8th Pt atom has been directly hypo-hyper-d-interelectronic bonded upon orderly distributed surface suboxide (oxygen vacant or vacancies) Ti atoms of Magneli phases. Smaller particle size in metal deposition so far has not been TEM recorded. However, while monoatomic particle distribution has been rather uniform and even, the Magneli phase surface has not been as uniformly covered by Pt monoatomic network. Polarization properties were exactly experimentally assessed on the basis of equivalent H-adatoms UPD estimated specific surface area and compared with so far the best Pt/Vulcan-XC-72 electrocatalyst (E-TEK, Somerset, NJ, 10 wt.% Pt) for the ORR (Fig. 6). Kinetically controlled current density values at potential of 0.85 V vs. RHE, where the mass transfer effect is negligible, were 0.33 mA cm-2Pt for pure Pt/C and 0.61 mA cm-2Pt for Magneli phases supported Pt (5.4 μg cm2 ). With about two order of magnitude increased Magneli phase specific surface area and one-to-one bonded Pt-Ti atoms, when the SMSI effect tends to its maximum, such outstanding results can certainly be further advanced. Three aspects of a remarkable electrocatalytic improvement can be accounted for: (a) Monoatomic dispersion implies dramatically increased electronic density of states along with the highest d-orbital strain (or Haruta named “forced”) effect, and consequently advanced catalytic activity; (b) The SMSI effect exponentially increases [29] with decreasing metallic catalyst particle size and approaches its maximum for monoatomic dispersion (Fig. 3, in ref. [29]); (c) The interactive Pt-OH spillover effect, which can not be unilaterally estimated besides the other two just considered


89

Figure. 7. Cyclic voltammograms scanned on Pt/(Magneli phases) 5.4 μg cm-2 (upper) and Pt/(Vulcan XC-72) (lower) in 0.5 M HClO4 t 297 K for different sweep rates (200, 100 and 50 mV s-1).

ones, in aqueous media and for Magneli phases is doubtless present and contributes to the overall catalytic properties. Under the same conditions Pt/C electrode, deprived from the spillover of primary oxide, shows remarkably poorer polarization characteristics for the ORR (Fig. 6). Such an overall picture additionally completes the comparison of cyclic voltammograms for these two electrodes, where Au/(Magneli phases)/Pt shows much broader potential range for the primary oxide (Pt-OH) growth with much higher charge capacities (Fig. 7). The controversial point so far has been that majority of papers dealing with the ORR unequivocally consider the first of four electrons transfer in the overall reaction of cathodic oxygen reduction for the RDS, but the primary oxide adsorption for the main obstacle and the rate determining factor [30-39]. First, although not obligatory, one would primarily expect some intermediate from the RDS reaction itself to appear as adsorptive rate determining species. The second, the reversible reaction of the primary oxide formation and its reversal reduction (eq. (1)) is extremely fast and at the reduction potentials for ORR proceeds quite independently and undisturbed along with the H-adatom anodic oxidation in LT PEM FC. Such state of experimental evidence is manifold confirmed by anodic oxidation of primarily formaldehyde and other aldehydes and monosaccharides within extremely broad potentiodynamic range [40]. In the same context, in order to check the value of the coverage with oxygen containing species in the range of potential close to the open circuit value, for Au/Ebonex/Pt (5.4 μg) electrode some specific additional potentiodynamic measurements were carried out. The electrode, previously hold at 0.20 V(RHE), was potentiostated at the selected initial potential in the range of potential characterized with lower Tafel slope on polycrystalline Pt (Fig. 6). The electrode was kept for 30 s at these potential, and then the potential

Figure. 8. Charge density that required oxygen species for their reduction, presented as a function of potential for Pt/(Magneli phases) (5 μg) electrode. The insert shows potentiodynamic I vs. E relations scanned from different initial potentials (hold) with sweep rate of 5 mV s-1. Where are Pt-OH species to impose the RDS?

was linearly scanned with 5 mV s-1 down to 0.0 V (RHE) (Fig. 8, insert). The amount of charge (Fig. 8) used in the scanning, determined by integration of I-E curves, proved that the surface coverage by Pt-OH on Au/(Magneli phases)/Pt (5μg) electrode was negligible and adsorption conditions were Langmuirian. The absence of PtOH layer, that supposedly inhibits oxygen reduction on polycrystalline Pt [30-39], is explained through the synergistic effect upon the catalyst, as a consequence of the interactions of Pt nanoparticles with Magneli phases, resulting in the spillover of Pt-OH and its spontaneous cathodic reduction [28]. The contradiction between the existing theory, based on the assumption that the primary oxide (M-OH), arises as the adsorptive intermediate decisive for the RDS in the ORR [3039], and the present experimntal evidnce, that an independent fast reversible reaction can by no means impose and keep any remarkable accumulative coverage at potentials of their maximal rates of consumption, implies the counterbalancing need to clear up the whole mechanism and eventually look for another chemisorptive intermediate of the true RDS. In other words, a simple kinetic analysis shows that the fast (and even independent) reversible reactions can by no means afford their rectants for the spillover accumulative intermediate adsorptive species (ΘM-OH → 1) decisive for another step to become the rate determining. 3.2. Au as the Reversible Hydrogen Electrode

90


Figure. 9. Cyclic voltammogram of polycrystalline Pt scanned in 0.1 M NaOH at sweep rate 100 mV s-1.

Gold is the noblest element amongst all transition metals [41], but the most inactive from the platinum group for hydrogen electrode reactions (HELR). Its position upon the asymmetric volcanic plots of individual electrocatalytic activity ( log jo ) for hydrogen evolution (HER) along the Periodic Table, after maximum at Pt, follows a rather sharp longer drop as compared with metals (W,Re,Os,Ir) within the ascending part of the curve [10,11]. Hydrogen and any atom or molecule with a filled one-electron level below the d-band gives rise to a repulsive coupling to the d-states for transition metals as long as the coupling matrix element is not strong enough as to push the antibonding peak above the Fermi level, so that Pt and Ni adsorb H-adatoms, while Au and Cu are deprived from its adsorption [41]. The question spontaneously imposes: Can we make Au less noble, H-adatom adsorptive upon it, and then becoming catalytically and electrocatalytically active like other Pt group metals [29,41-43]? The reversible hydrogen electrode behavior primarily implies the reversible H-adatoms adsorption and desorption as marked by their equivalent and symmetric charge capacities upon cyclic voltammogram. Although gold with completely filled d-orbitals at the ground state represents the post-transition element and thus the noblest one [41-43], in its metallic lattice Au still has one empty “d-seat”, and thereby still adsorptive d-band [6]. The early and majority of other papers [44,45] deny any remarkable adsorption, while there also exists a lot of convincing experimental evidence [46-51] that Au both adsorbs and even absorbs a great deal of hydrogen on its surface and its bulk, respectively. In such a respect, Costa et al. [46-48] have scanned a discernible peak of hydrogen diffusional desorption, as well as two distinct transition times for galvanostatic oxidation of atomic hydrogen, one for diffusive desorption of H-abatoms from the bulk of metal lattice and another for adsorbed H-adatoms, though they failed to record its corresponding mutual peak of adsorption simultaneously accompanied by absorption. Arvia

Figure. 10. Cyclic voltammograms of spectroscopically pure Au stepwise scanned towards the oxygen evolving limit in alkaline heavy water solution (0.1 M NaOD, D2O) with sweep rate of 200 mV s-1.

et al. [49-51] associate the gold/aqueous acidic interface activation for hydrogen electrooxidation with potentiodynamic aging, which implies surface restructuring by the electrosorption and electrodesorption of oxygen on the metal surface (anodic surface oxide growth, Au=O, and its removal during the cathodic scans). Even Pt does not feature any remarkable adsorption of Hadatoms when potentiodynamic scans are restricted within cathodic hydrogen evolution and double layer charging range [52-54]. Full UPD coverage (ΘH → 1) Pt approaches and even exceeds (OPD) only inside completed voltammograms [55], which includes oxygen evolving limits (Fig. 9), and thereby the entire electrode surface restructuration. In such a context, it is generally well known in heterogeneous catalysis that maximal surface area, and thereby maximal catalytic activity enables (provides) first the transference of metallic salts into oxides with subsequent reduction in hydrogen furnace to a proper nanostructured metallic catalyst state, while direct reduction in hydrogen stream shrinks catalytic particles. Typical issue illustrates a mild thermal decomposition of metallic acteyilacetonates (2,4-pentanedionates) of noble metals upon anatase titania and other hypo-delectronic oxide supports as composite catalysts [4,5]. In such a respect, potentiodynamic investigations with spectroscopically pure Au in alkaline heavy water solution (0.1 M NaOD, D2O) have been of particular interest, since deuterium has both twice larger mass and radius as compared with protium, and its effect upon the restructuring of electrode surface should have to be remarkable and more pronounced. In other words, cyclic alternation between AuOD, Au=O and Au-D implies stepwise penetration of absorbed deuterium atoms deeper in Au metal lattice.


Figure. 11. The same voltammograms of Au in alkaline heavy water scanned after multifold cyclization and established reversible properties of hydrogen electrode, when surface restructuration is at its steady state and further hydrogen adsorption and desorption do not depend on the complete voltammogram recycling up to anodic oxygen evolution.

In such a respect, initial potential cyclisation scans between hydrogen evolving limits and double layer charging range clearly show the appearance of the small reversible UPD peaks for hydrogen adsorption and desorption (Fig. 10). The charge capacity of the latter does not change with cyclization and persists to be rather limited in H-adatoms coverage. It should be noticed that the initial UPD of H-adatoms on Au arises far from the hydrogen evolving limits. Further stepwise extension of potential limits towards anodic oxygen evolution reveals both the UPD and diffusional desorption peaks of remarkably larger and successively growing charge capacity, well separated along the potential axis and resembling to similar voltammogram shape and behavior of Pd. Namely, the anodic hydrogen oxidation peak scanned on Pd primarily characterizes prevailing diffusional, and thereby postponed desorption along potential axis, since the latter takes place from the bulk of metallic lattice. The restructuring of Au surface as imposed by the effect of successive anodic oxide growth apparently becomes even more pronounced by the following stepwise deuterium penetration in the bulk of metal lattice and therefore, deeper and deeper next restructuration influenced by Au=O. Anodic part of the primary (Au-OH) and surface oxides (Au=O) growth is typical for Au, though two desorption peaks of the latter are more distinctly separated and almost of the same charge capacity. Repeated cyclization at every stepwise extending limit towards oxygen evolution shows the initial jump and further stabilization of the final steady-state voltammogram, that way indicating (pointing to) the restructuring effect of grow-

91

ing oxide deposits. Successive cyclization of the whole voltammogram scanned between hydrogen and oxygen evolving limits, shows that as a progressive effect of altering deuterium adsorption (and absorption) and desorption, along with the oxide restructuring effect, these two peaks tend to each other along the potential axis, to establish their reversible relation. As the consequence, hydrogen evolving limit becomes successively shifted for about 400 mV from its initial towards more positive potentials, and finally gets established the thermodynamic equilibrium at the reversible Nernst potential value (Fig. 11, a & b). The voltammogram in the hydrogen adsorption, desorption and evolving range resembles now more to Pt than Pd, and reveals its final reversible properties of hydrogen electrode. Once when Au finally earns such reversible properties after manifold cyclization and thorough restructuring, its reversible properties become further continuously maintained even within narrow potential range associated only with hydrogen adsorption, desorption and cathodic evolution (Fig. 11a). Such an Au electrode when straight transferred in the same alkaline solution of regular water, repeats the same potentiodynamic spectra and features with protium as with deuterium. However, if shortly treated in flame to remove the achieved restructuring properties, such an electrode reveals potentiodynamic spectra characteristic for Au bulk of metal in normal (regular) aqueous media. Meanwhile, everything repeats again, as just displayed, whenever Au proceeds the same repeatable potentiodynamic cyclisation in heavy water. In the same context, one should be aware that even the true hydrogen reversible electrode implies a rather developed platinized Pt surface and to be protected from oxygen contact. 3.3. SMSI of Gold Upon Anatase Titania The question thus naturally imposes: can we provide continuous properties of the reversible hydrogen electrode for Au in the steady-state conditions of Low Temperature PEM Fuel Cells and other practical applications? While neither Au, nor titania by themselves are individually active for CO and O2 adsorption, nanostructured Au/TiO2 oxidizes CO even much below room temperature [42,56-61]. In the light of the existing electrocatalytic theory [1-5], such an effect could a priori be expected, if the original Brewer interatomic bonding theory, with all accompanying consequences [6-9] , holds for the interactive interionic bonding, too. In other words, massive Au and nanostructured supported Au/TiO2, or Pt vs. Pt/TiO2, in their catalytic features are distinctly different species. Haruta et al. [62] have shown that the same reactants (propylene along with equimolar mixture of hydrogen and oxygen) yield different products upon different nano-sized Au catalyst supported on anatase titania (Au/TiO2): (i) Propane by hydrogenation at nano-particles < 2nm Au, and (ii) Propylene oxide by epoxidation or oxygen addition for > 2nm Au. Hydrogenation implies H-adatom adsorption upon Au, that never spontaneously occurs on pure massive bulky gold surface. Haruta [42,56-62] ascribes such chemisorptive prop-


92

Au-OH

Au 4f

Au foil

XPS Intensity /a.u.

voltammetry. The deconvoluted Au 4f peaks with lower Au loadings reveal that Au nanoparticles in interactive bonding contact with titania appear partially oxidized [26]. The peak located at 82.15 ± 0.1 eV is identified and attributed to metallic Au, while the peak at 84.05 ± 0.1 eV is situated for 1.9 eV towards higher binding energy and corresponds to the primary (Au-OH or AuOOH) oxides. The latter, in accordance with the present theory, appear as the a priori provided the primary oxide spillover species, associated with anatase titania interaction (eq. (1)), and are in advance, already available for anodic CO oxidation.

1.6 eV

2

0.4 μg Au/cm

2

0.8 μg Au/cm

2

1.6 μg Au/cm

2

2 μg Au/cm

-92

-90

-88

-86

-84

-82

Binding Energy / eV

-80

-78

-76

Figure. 12. XP spectra of Au 4f for vapor deposited nanostructured Au upon a fine thin film of anatase titania with deconvolution for lower amounts of deposits to reveal the existence of primary oxides (Au-OH an AuOOH) [26].

erties to “forced Au-d-orbitals”, or strained in our terminology, in particular when SMSI deposited on anatase titania. In such a respect, as already stated above, there have been investigated rather fine Au films as deposited by electron beam evaporation from an Au wire of ultrahigh purity under vacuum onto stationary titania coated microscopical slides [25]. The XP spectra of the Au 4f electrons reveal the remarkable binding shift (Fig. 12), that testifies for the SMSI on the interphase Au/TiO2 and this is one of the first such evidence in heterogeneous catalysis [26]. The smaller the nanoparticle size of Au deposit, as theoretically predicted in the present paper, the larger the binding energy shift in XP spectra with titania, and the more pronounced arises the SMSI effect, with tendency to its maximal binding effect at monoatomic dispersion and 1:1 deposition of Au on Ti atoms. Such Mavrikakis [29] type of trend in the bonding effectiveness and consequently the higher Haruta [56-61] type catalytic activity, lead us to develop the reversible hydrogen electrode, substantiated on highly dispersed subnanostructured Au SMSI upon anatase titania and/or Magneli phases [63], in particular when their interphase located oxides become mutually reduced to yield the intermetallic phase TiAu4. The whole rather pronounced electrocatalytic and reversible electrode properties result, alike the monoatomic Pt upon Magneli phases, as the effect of its strained dorbitals due to a rather high dispersion of nano-particulate Au and the strong interactive bonding effectiveness (SMSI) upon the hypo-d-electronic support. The latter corresponds and arises as even the more pronounced effect of d-orbital strain expansion in such a metal/support d-d-bonding effectiveness, relative to the Au surface restructuration in cyclic

3.4. Thermodynamic Basis of Hypo-Hyper-dInteratomic Bonding What are the thermodynamic roots of the moisture effect in Haruta [56-61] terminology as decisive for heterogeneous SMSI catalytic processes, to occur or not? What does it mean the oxophilicity of hypo-d-electronic transition metals (Fig. 1 in ref. [64]) in correlation with the interactive spillover effect of the primary oxides, as usually being spontaneously induced by dissociative adsorption of water molecules upon hypo-doxide supports of composite (electro)catalysts? The point is that hypo-d-electronic transition metals are rather oxophilic elements and thereby subject of specific and tedious metallurgical production, while there affinity for intermetallic bonding with hyper-d-metals is pronouncedly high and usually results even in extraordinarily stable Laves type intermetallic phases [6-9]. The individual d-dinteratomic cohesive bonding increases from d1 to d5 level with its maximum at the peaks of volcano plots along three transition series [1,2,6-9]. Meanwhile, the hypo-hyper-d-dbonding effectiveness follows the volcano plot along each semi-d-series in both combinations, a hyper-d-element while alloying along hypo-d-semi-series, and vice versa. The latter is accompanied by unusually high negative enthalpies of formation and, as a consequence, dramatically increased their intermetallic melting points, and further consequently, by a rather pronounced synergism in electrocatalysis for the HELR [1-9,12] (Fig. 1). In such a respect, one should notice that two maximum values of individual transition metal plots does not coincide for cohesive bonding (d5) [1] and electrocatalytic activity in hydrogen evolution (d8) [1,3], while they fully overlap for hypo-hyper-d-interelectronic combinations (at average d8) [12]. The main breakthrough approach of the present method of composite catalysts formation primarily consists in a proper application of the extended Brewer interionic bonding theory [1-5], to use the hypo-hyper-d-interionic d-d-bonding effect at molecular level of properly mixed metal ions and/or molecules, for example, their oxides or hydroxides in a consistent stoichiometric ratio, and reduce the overall reaction on the ‘neutralization’ or water production (steam withdrawal), such as TiO2 + 3 Pt + 2 H2 = TiPt3 + 2 H2O

(2)

In other words, while as the effect of rather pronounced


93

Table I. Relevant thermodynamic values for present calculations (after Kubichewski et al.28).

ΔΗ298 (ΚJmol-1)

S298 (Jmol-1 K-1)

TiO2 (s) (rutile)

-944.0

50.6

TiO (s) (monocl.)

-542.7

34.7

HfO2 (s) (monocl.)

-1117.5

59.4

ZrO2 (monocl.)

-1100.8

50.4

H2 (g)

0

130.6

Ti (s)

0

30.7

-241.8

188.7

Sample

H2O (g)

oxophilicity, there is no real technological ability to reduce individually titania, hafnia, zirconia, ceria, lanthania, etc., at reasonably and attainably high temperatures to their metallic state (Appendix Fig. 1A), the extended Brewer d-dinteractive method enables to combine them properly and substantiate the entire desirable intermetallic phase formation within the simultaneous reaction with Pt, Pd, Ir, Ni, Co, or their oxides (Eq. (2)) and/or other salts (preferably nitrates, acetates, oxalates and similar combinations, because of their simple thermal decomposition), at only few hundreds degree of centigrade [5,6]. In such a context, the extraordinarily high enthalpies of formation of the Brewer type stable Laves phases, as a thermodynamic driving force and its counteracting effect against the oxophilicity of individual hypo-delements, succeed in a greater amount to annul (subtract) the prevailing part of heats necessary to reduce hypo-d- and/or hypo-f-oxides and hydroxides in such a properly composed overall reaction. The main theoretical impact to establish a novel nanotechnology of broad and specific field of hypo-hyper-d-interionic and interatomic catalysts has been based on the interactive reduction, in which atoms of Pt, Pd, Co, Ni, and/or any other suitable hyper-d-electronic transition metal or their salts, behave both as the reactant to produce defined intermetallic phase (TiPt3, HfNi3, ZrPd3, CeCo5, LaNi5, etc.), and as a self-catalyst to reduce hypo-d- or hypo-f-oxides, like TiO2, HfO2, ZrO2, CeO2, LaO2, in course of their mutual interaction, so that the overall reaction, (Eq. (1)), can be thermodynamically considered to occur as split in two steps, consisting from the straight, catalytically facilitated (Pt) titania reduction,

Figure. 13. Temperature dependence of various hypo-doxides (TiO, TiO2, ZrO2, HfO2) reduction to their individual metallic state, and catalytic formation of their Laves type stable intermetallic phases (TiPt3, ZrPt3, HfPt3), as a function of parametric ratio of hydrogen and water vapour partial pressure.

Since thermodynamic functions and data are additive within defined stoichiometry (extended Hess law, Haber-Born cycle), from the existing literature one simply has as follows, ΔG1 (TiO2→ TiPt3) = ΔG1-1 +

ΔG1-2 = - RT ln K1 =

= (460,400 – 96.3 T) + (- 341,833 + 33.51 T)

(3)

The point and substance is in a huge free enthalpy of the Brewer type intermetallic hypo-hyper-d-interelectronic phase formation, that succeeds energetically (or, catalytically) to annul (subtract) great deal of TiO2 stability. When one now introduces the equilibria constant in the overall dependence, which practically reflects in the ratio of partial pressures of hydrogen and water vapor, other reactants being at their standard states, there follows a parametric relation, ΔG1(TiO2→TiPt3) = 118,567 — 62.79 T pH 2

= 38.29 T log ( p ) H O

(4)

2

TiO2 + 2 H2 = Ti + 2 H2O

(2-1)

and further direct Brewer intermetallic bonding reaction, Ti + 3 Pt = TiPt3

(2-2)

The reliable relevant thermodynamic functions for this type of reaction of Pt with Ti, Zr and Hf, we owe to Meschter and Worrell [65,66], while the broad thermodynamic survey of data for oxides afford Kubaschewski at al. [67], (Table I).

When the latter is compared with the individual thermodynamic equation for TiO2 reduction [67], ΔG (TiO2) = 460,400 — 96.3 T pH 2

= 38.29 T log ( p ) H O 2

(5)


94

one has now a rational thermodynamic basis and tools to compare temperature (T) both of individual titania, zirconia and hafnia reduction, with the overall Brewer effect of mutual interactive reduction, for three such systems TiPt3, ZrPt3, HfPt3, along the partial pressure relation, taken as an axis (Fig. 13). While titania itself thermodynamically requires, for example, minimum of 2180 K to be reduced at simple technical conditions (

pH 2 p H 2O

=1,000), under the same

circumstances, there follows TiPt3 formation at 667 K. Although one can provide hydrogen supply even with less than 1 ppm of moisture, this graph is restricted at technically accessible real conditions, since in the course of such processes the main accompanying product is deliberated water. On the electrochemical Volta scale, just for the comparison purposes to assess the Brewer interactive effect contribution, these two reactions would thermodynamically occur in their spontaneous directions at 1.1184 and 0.2587 V, respectively, and such remarkable difference favours the appearance and growth of strongly bonding Laves type of Brewer intermetallic phases. In other words, while rather strong reducing agencies (like NaBH4) are absolutely unable to reduce individually titania, hafnia, or zirconia, their stoichiometric molecular mixtures with Pt, Pd, Ni salts, resulting in Laves type intermetallic phases, are accordingly, attainable (accessible), at least partially. In such a context it is worthwhile noting that in agreement with the present theory, catalytic literature shows usual (interactive) annealing at about 300 - 400 oC for all intermetallic catalysts, though it is much below their individual melting points. The same type of corresponding thermodynamic relations for Hf and Zr are as follows, respectively [65-67]: ΔG (HfO2→HfPt3) = (633,900 — 100.3 T) + (-475,302 + 52.30 T) = ⎛ p

⎞

H = 158,598 — 48.0 T (J mol-1) = 38.29 T log ⎜⎜ p ⎟⎟ ⎝ HO⎠ 2

(6)

2

ΔG (ZrO2→ZrPt3) = (617,200 — 104.8 T) + ( -452,374 + 52.72 T) = ⎛ pH 2 ⎞

= 164,826 — 52.08 T (J mol-1) = 38.29 T log ⎜⎜ p ⎟⎟ H O ⎝

2

⎠

(7)

This is the substance and basis for a novel nanotechnology of hypo-hyper-d-intermetallic nano-particulate cluster phase formation: the stronger the intermetallic d-d-bonding, the lower is the mutual thermodynamic temperature of intermetallic phase reduction. The bonding effectiveness apparently increases with the exposure of d-orbitals (Fig. 13), from 3d towards 5d level, or from Ti towards Hf, and this way confirms the initial basic statements of the Brewer intermetallic bonding theory [6-9]. The broad common point is that all hypo-hyper-d- and/or hypo-f-hyper-d-intermetallic phases characterizes pronounced strong interionic (or interatomic)

bonding effectiveness, and this implies the versatile and broad number of combinations by simple following of the present technology and thermal procedure. The same thermodynamic calculations based on lower oxide states of TiO, ZrO and HfO, as the starting reactants, show that in contact with Pt they are absolutely unstable and instantaneously produce stable TiPt3, ZrPt3 and HfPt3 intermetallic phases. Such fact is noteworthy, since supported catalysts (M/TiO2) immediately undergo a rather facilitated reduction of titania in hydrogen furnace or even by adsorbed H-adatoms (M-H) during heterogeneous catalytic process, to a non-stoichiometric state (TiOx, where 2 > x > 1) named «decoration» or «encapsulation» [16-18], that finally results in either in Magneli phase (TinO2n-1) [63] or in defined intermetallic phases [4,5]. This is the Brewer High Temperature Thermodynamics and there is no remark or reproach. However, kinetics is something else and still requires further thorough studies. These processes with Ti, Zr, Hf hardly proceed to their completeness, so to be considered fully quantitative, in particular upon highly oxophilic hypo-d-oxide supports. They require other specific catalysts, crystallization agencies and more specific procedure primarily as a time function. Oxides of less oxophilic metals and thereby of lower reduction temperatures, like Mo, W, Cr, V, when combined with Ni, Co, Fe, or their salts, follow the same type of Brewer reactions at much lower temperatures [4,5]. As thermal gravimetry analysis (TG) shows [4,5], such reactions proceed at dramatically lower temperature relative to individual oxides of oxophilic and refractory metal, and follow the exact stoichiometry. When stated the hypo-hyper-d-interionic bonding effect, this imples, for example, hyper-d-electronic (RuO2,IrO2, individual or mixed) oxides upon Magneli phases and/or anatase titana, or hypo-d-oxide species, as advanced composite SMSI electrocatalysts for anodic oxygen evolution (OER). The aim of the present paper has been to afford thermodynamic bases for the Brewer type intermetallic phase formation out of oxophilic hypo-d-electronic oxides, as the effect of enormous enthalpies accompanying all hypo-hyper-dinterelectronic interactions of transition metals and their ions. In other words, the attainable conditions to produce extraordinary stable hypo-hyper-d-interelectronic phases out of hypo-d-oxides result as a direct consequence of the Brewer intermetallic bonding theory [1-9]. The sound proof of such thermodynamic analysis for a direct thermal production of extraordinary stable hypohyper-d-intermetallic phases of transition metals, affords the spontaneous growth of a Laves type stoichiometric compound, TiPt3, at the phase boundary of Pt/TiO2 (Fig. 3). Namely, H-adatoms from Pt surface, step by step, succeed to produce XPS confirmed TiPt3 intermetallic phase even at lower temperature, about 300 oC, than it has been thermodynamically predicted. The point is that thermodynamic calculation implies molecular hydrogen, while one here deals with chemisorbed H-adatoms, which dissociative adsorption upon Pt surface proceeds spontaneously and appear energetically more predestined for the facilitated reduction reaction. A rather specific issue of interactive hypo-d-oxide sup-


ported MoPt3/TiO2/C electrocatalyst has recently been thoroughly investigated, surface charcterized and displayed in details elsewhere in the light of the present theoretical model [68].

4.0. Concluding Remarks This paper has shown that Magneli phases represent an unique interactive support for composite electrocatalysts in LT PEM FC, water electrolysis and other applications in broader electrochemical practice. With their (a) rather high metallic type electron conductivity, Magneli phases are the best current collecting and an outstanding supporting material, (b) the strong and orderly hypo-hyper-d-interelectron bonding and consequently, transcendentally SMSI inducing less H-adatom adsorptive strengthening of such an intermediate upon metallic part of catalyst, (c) advanced spillover interactive hypo-d-oxide supports for primary oxides (M-OH) transference and (d) by their surface chemical structure, a priori ready and predestined for an ordered monoatomic network dispersion in individual or prevailing hyper-delectronic catalyst grafting. As an extremely orderly distributed layerlike structure, in particular for suboxide reacting centers network distribution, Magneli phases of rather high chemical stability and almost metallic type conductivity, appear as such an unique and highly advanced, almost unrepeatable catalytic interactive support for monoatomic (SMSI) network dispersion of metallic part of electrocatalysts. The weak point of Magneli phases at the present time is their at least for two order of magnitudes larger particle size and consequently smaller available surface area for interactive bonding of catalyst network. A direct reduction of nanostructured anatase titania colloidal particles is promising to enable resolving such a problem. Nanostructured Magneli phases might provide homogeneous monoatomic inter-bonding dispersion network of Pt (1:1 of Pt versus surface oxygen vacant suboxide Ti atoms) and increase the catalytic activity for the ORR at least for an order of magnitude, that already is and would have further been a great result from theoretical point of view, and will be shown elsewhere, too [69]. Meanwhile, LT PEM FC for car traction could by no means rely of noble metals of rather limited earth sources. What we really need for broader application of hydrogen energy are alkaline, OH- - ions conductive membranes, operating at higher temperatures (approaching 200 oC), when the cell voltage becomes more determined by the thermal effect, and less so by electrode material or electrocatalysis, and the entire phenomenon arises primarily thermodynamically determined. Then noble metals might be replaced by suitable and cheaper transition metals and specific compounds. The present paper has shown that highly nanodispersed and (SMSI) titania (or Magneli phases) supported Au can be the best CO tolerance and even the reversible hydrogen electrode, a priori rather good for the ORR, too, but the same remark imposes like for Pt and all other noble metals in broader application: it is rather expensive and even more so, rather limited by the existing earth sources.

95

5.0. Acknowledgements This paper is dedicated to distinguished Professor Helmut Boennemann, who has been the main creator of nanostructured colloidal precursors in contemporary electrocatalysis, for his 65th anniversary. The present paper has been supported by and carried out within EU Project ‘Apollon’, Contract Nr.ENK5-CT-200100572, EU Project NR. NNES-2001-00187, and the Project ‘Prometheas’, Contract Nr. ICA2-2001-10037. REFERENCES [1] M. M. Jaksic, J. New Mat. Electrochem. Systems, 3 (2000) 153. [2] M. M. Jaksic, J. Mol. Catal. 28 (1986) 161. [3] M. M. Jaksic, M. M. Electrochim. Acta, 45 (2000) 4085. [4] S. G. Neophytides, S. Zafeiratos, M. M. Jaksic, J. Electrochem. Soc., 150 (2003) E512. [5] S. G. Neophytides, S. Zafeiratos, G. D. Papakonstantinou, J. M. Jaksic, F. E. Paloukis, M. M. Jaksic, Int. J. Hydrogen Energy, 30 (2005) 131, 393. [6] L. Brewer, Science, 161 (1968) 115. [7] L. Brewer, in: Electronic Structure and Alloy Chemistry of Transition Elements, Beck, P. A., Ed.; Interscience, New York, 1963, p. 221. [8] L. Brewer, in High-Strength Materials, Wiley, Zackay, V. F., Ed.; Wiley New York, 1965, p. 12; L. Brewer, The Cohesive Energies of the Elements, LBL-3720, Berkeley, California, May 4, 1977. [9] L. Brewer, In Phase Stability in Metals and Alloys, Rudman, P.; Stringer, J.; Haffee, R. I., Eds.; McGraw-Hill, New York, 1967, p. 39. [10]H. Kita, J. Elctrochem. Soc. 113 (1966) 1095. [11]M. H. Miles, J. Electroanal. Chem., 60 (1975) 89. [12]M. M. Jaksic, C. M. Lacnjevac, B. N. Grgur, N. V. Krstajic, J. New Mat. Electrochem. Systems, 3 (2000) 169. [13]S. J. Tauster, S. C. Fung, J. Catalysis, 55 (1978) 29. [14]S. J. Tauster, S. C. Fung, R. T. K. Baker, J. A. Horsley, Science, 217 (1981) 217, 1121. [15]S. J. Tauster, S. C. Fung, R. L. Garten, J. Aam. Chem. Soc., 100 (1978) 170. [16]S. A. Stevenson, Metal-Support Interactions in Catalysis, Sintering and Redispersion, Van Nostrand, New York, 1987. [17]L. L. Hegedus, Ed., Catalysis Design, Wiley, New York, 1987. [18]G. L. Haller, D. E. Resasco, Adv. Catal. 36 (1989) 173. [19]P. Sabatier, La Catalyse en Chimie Organique, Librairie Polytechnique, Paris, 1913; P. Sabatier, Ber. Deutsche Chem. Soc., 44 (1911) 2001. [20]A. Vittadini, A. Selloni, F. P. Rotzinger, M. Gratzel, Phys. Rev. Lett., 81 (1998) 2954.

96


[21]M. Lazzeri, A. Vittadini, A. Selloni, Phys. Rev. B, 63 (2001) 155409. [22]A. Tilocca, A. Selloni, J. Chem. Phys, 119 (2003) 7445. [23]J. Livage, M. Henry, C. Sanchez, Prog. Solid State Chem., 18 (1988) 259. [24]W. Li, C. Liang, W. Zhou, J. Qiu, Z. Zhou, G. Sun, Q. Xin, J. Phys. Chem. B, 107 (2003) 6292. [25]I. M. Arbatzis, S. Antonaraki, T. Stergiopoulos, A. Hiskia, E. Papaconstantinou, M. C. Bernard, P. Falaras, J. Photochem. Photobiol. A, 149 (2002) 237. [26]I. M. Arbatzis, T. Stergiopoulos, D. Andreeva, S. Kitova, S. G. Neophytides, P. Falaras, J. Catalysis, 220 (2003) 127. [27]S. Zafeiratos, G. Papakonstantinou, M. M. Jaksic, S. G. Neophytides, J. Catalysis, 232 (2005) 127. [28]Lj. M. Vracar, N. V. Krstajic, V. R. Radmilovic, M. M. Jaksic, J.Electroanal. Chem., 587 (2006) 99. [29]M. Mavrikakis, P. Stoltze, J. K. Norskov, Catal. Lett., 64 (2000) 101. [30]S. Goetesfeld, T. A. Zawodzinski, in Advances in Electrochemical Science and Engineering, R. C. Alkire, D. M. Kolb, Eds., Wiley-WXH, Weinheim, Vol. 5, pp. 197-297, 1997. [31]A. Anderson, Electrochim. Acta, 47 (2002) 3759. [32]N. M. Markovic, H. A. Gasteiger, P. N. Ross, J. Phys. Chem.,16 (1996) 6715. [33]S. Mukerjee, S. Srinivasan, in W. Vielstich, A. Lamm, H. Gasteiger (Eds.), Handbook of Fuel Cells, Vol.2, John Wiley & Sans. Inc., Chichester, UK, pp 502-519, 2003. [34]R. R. Adzic, in Frontiers in Electrochemistry, Vol. 5, Electrocatalysis, J. Lipkowski, P. N. Ross, Eds., VCH Publishers, New York, p. 197, 1998; N. A. Anastasijevic, V. B.Vesovic, R. R. Adzic, J. Electroanal. Chem. 229 (1987) 317. [35]M. R. Tarasevich, Sadkowski, E. Yeager, in Comprehensive Treatise of Electrochemistry, B. E. Conway, J. O'M. Bockris, E. Yeager, S. U. M. Kahn, R. E. White, Eds., Plenum Press, New York, p. 301, 1983. [36]D. B. Sepa, M. Vojnovic, A.Damjanovic, Electrochim. Acta, 26 (1981) 781. [37]A. Damjanovic B. B. Sepa, M. V. Vojnovic, Lj. Vracar, A. Damjanovic, Electrochim. Acta, 31 (1986) 91. [38]A.Damjanovic, A. Dey, J. O'M. Bockris, Electrochim. Acta, 11 (1966) 791. [39]J. O'M. Bockris, R. Abdu, J. Electroanal. Chem., 448 (1998) 189; J. O'M. Bockris, A. Damjanovic, J. McHardy,

C.R. Troisieme Journ. lnternat. d'Etude des Piles a Combustible, Presses Academique Europeenne, Bruxelles, pp. 15-28, 1969. [40]N. Ristic, M. Kotorcevic, C. M. Lacnjevac, A. M. Jokic, M. M. Jaksic, Electrochim. Acta, 45 (2000)2973. [41]B. Hammer, J. K. Norskov, Nature, 376, 238 (1995). [42]M. Haruta, The Chemical Record, 3 (2003) 75.

[43]M. Valden, X. Lai, D. W. Goodman, Science, 281 (1998) 1647. [44]M. Breiter, C. A. Knorr, V. Volkl, Ber. Bunsenges. Phys. Chem., 59 (1955) 681. [45]F. G. Will, C. A. Knorr, Ber. Bunsenges. Phys. Chem., 62 (1960) 378. [46] F. Chao, M. Costa, C. R. Ser. C 284 (1977) 603, 763. [47]F. Chao, M. Costa, P. Elkaim, C. R. Acad. Acad. Sci. (Paris), Sci. (Paris), Ser. C 284 (1977) 639. [48]F. Chao, M. Costa, R. Parsons, C. Grattepain, J. Electroanal. Chem., 115 (1980) 31. [49]R. Cordova Orellana, M. E. Martins, A. J. Arvia, Electrochim. Acta, 24 (1979) 469. [50]R. Cordova Orellana, M. E. Martins, A. J. Arvia, J. Electrochem. Soc., 127 (1980) 2628. [51]M. E. Martins, J. J. Podesta, A. J. Arvia, Electrochim. Acta, 32 (1987) 1013. [52]W. E. O’Grady, M. Y. C. Woo, P. L. Hagens, E. Yeager, J. Vac. Sci. Technol., 14 (1977) 365. [53]E. Yeager, W. E. O’Grady, M. Y. C. Woo, P. L. Hagens, J. Electrochem. Soc., 125 (1978) 348. [54]P. N. Ross, Surf. Sci., 102 (1983) 141. [55]J. Clavilier, in Interfacial Electrochemistry, Theory, Experiment, and Applications, A. Wieckowski, Ed., p. 231, Marcel Dekker, 1999. [56]M. Haruta, M., Catal. Today, 1997, 36, 153. [57]Date, M.; Haruta, M. J. Catalysis, 201 (2001) 221. [58]Boccuzzi, F., Chiorino, A.; Tsubota, S.; Haruta, M. J. Catalysis, 202 (2001) 256. [59]F. Boccuzzi, A. Chiorino, M. Manzoli, P. Lu, T. Akita, S. Ichikawa, M. Haruta, J. Catalysis, 202 (2001) 256. [60]T. Akita, P. Lu, S. Ichikawa, K. Tanaka, M. Haruta, Surf. Interface Anal., 31 (2001) 73. [61]M. Haruta, Catalysis Survey Japan, 1 (1997) 61. [62]T. Akita, K. Tanaka, S. Tsubota, M. Haruta, J. Electron Microscopy, 49 (2000) 657. [63]A. Magneli, Acta Chem. Scand., 13 (1959) 5. [64]K. L. Ley, R. Liu, C. Pu, Q. Fan, N. Leyarovska, C. Segre, E. S. Smotkin, J. Electrochem. Soc., 144 (1997) 1543. [65]P. J. Meschter, W. L. Worrell, Metallurgical Trans. A, 7 (1977) 299. [66]P. J. Meschter, W. L. Worrell, Metallurgical Trans. A, 8 (1977) 503. [67]O. Kubaschewski, C. B. Alcock, P. J. Spenser, Materials Thermochemistry, 6th ed., Pergamon Press, Oxford, 1993. [68]J. M. Jaksic, Lj. M. Vracar, S. G. Neophytides, S. Zafeiratos, G. Papakonstantinou, N. V. Krstajic, M. M. Jaksic, Surf. Sci., 598 (2005) 156. [69]N. V. Krstajic at al., to be published


97

APPENDIX

Fig. 1A. Nomographic parametric presentation of thermodynamic temperature depending behaviour of various oxides as a function of H2/H2O, CO/CO2 and O2 ratio or content [67].